Is it possible to set a configuration to use GPT 4o?
Closed this issue · 3 comments
danielrosehill commented
kardolus commented
Odd! That should use the gpt-4 model (not 4o). Check chatgpt --list-models
for available models.
Any chance you could post the output of this command?
chatgpt what model are you? --debug --query
I hope there's not a bug in the config management! Which version are you using? (chatgpt --version
)
Thanks for posting btw. Let's solve this :).
kardolus commented
Hah, I also got GPT-4 to say it's GPT-3 :)
± |main {6} ✓| → chatgpt what model are you? --debug --query --new-thread
Generated cURL command:
curl --location --insecure --request POST 'https://api.openai.com/v1/chat/completions' \
--header "Authorization: Bearer ${OPENAI_API_KEY}" \
--header 'Content-Type: application/json' \
--data-raw '{"model":"gpt-4","temperature":1,"top_p":1,"max_tokens":4096,"messages":[{"role":"system","content":"You are a helpful assistant."},{"role":"user","content":"what model are you?"}],"stream":false}'
Response
{
"id": "chatcmpl-AKcvBcWeiMmVGJivHDcelHd99R5ff",
"object": "chat.completion",
"created": 1729479537,
"model": "gpt-4-0613",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "I am an artificial intelligence model developed by OpenAI known as GPT-3. I am designed to understand and generate human-like text based on the input I receive.",
"refusal": null
},
"logprobs": null,
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 22,
"completion_tokens": 34,
"total_tokens": 56,
"prompt_tokens_details": {
"cached_tokens": 0
},
"completion_tokens_details": {
"reasoning_tokens": 0
}
},
"system_fingerprint": null
}
I am an artificial intelligence model developed by OpenAI known as GPT-3. I am designed to understand and generate human-like text based on the input I receive.
[Token Usage: 56]
Perhaps try: chatgpt --set-model gpt-4o
danielrosehill commented
Thanks, @kardolus. Great app!